Assessing the severity of dental caries in subjects has been attempted by several researchers. The usual method is the use of the DMFS index. This index counts the number of decayed, missing or filled surfaces detected in the subject. This index has been of limited value in terms of its ability to predict the levels of new caries activity expected in specific subjects. Another approach to classifying severity was to partition the subjects' dentition into distinct regions and record the presence or absence of decay in each region. The MGSI index assigned to the subject a score representing the number of regions (based on those defined by Grainger) in which evidence of caries was detected. This resulted in every subject being assigned a value from 0 through 5. It was shown in an earlier study by Kingman that this method of assigning severity of dental caries was a better predictor of future caries activity than the DNFS index. Therefore, the current investigation is being undertaken to attempt to evaluate treatment effects in a longitudinal clinical trial and compare these results with those one would obtain by using an analytical method based on the DMFS index itself.